Search

CN-121993008-A - Multi-mode perception-based vehicle door anti-collision decision method and system

CN121993008ACN 121993008 ACN121993008 ACN 121993008ACN-121993008-A

Abstract

The invention provides a vehicle door anti-collision decision method and system based on multi-mode perception, comprising the steps of obtaining a door lock state of a target vehicle door, and gear information and state information of a vehicle; and determining a risk level based on the obstacle information, and generating a control instruction of the target door based on the risk level and the intention of the occupant in the vehicle. The intelligent degree of the collision avoidance decision of the vehicle door is improved, the safety is improved, and unnecessary risks caused by the collision of the vehicle door are avoided.

Inventors

  • ZHAO BEIBEI

Assignees

  • 安徽开阳科技有限公司
  • 奇瑞汽车股份有限公司

Dates

Publication Date
20260508
Application Date
20260324

Claims (10)

  1. 1. The vehicle door anti-collision decision method based on multi-mode perception is characterized by comprising the following steps of: Acquiring the door lock state of a target door and gear information and state information of a vehicle; If the target vehicle door is in an unlocking state, the gear information of the vehicle is a target gear, and the vehicle is in a parking state, the information of an obstacle outside the vehicle and the intention of an occupant in the vehicle are acquired; A risk level is determined based on the obstacle information, and a control instruction for the target door is generated based on the risk level and the intention of the occupant in the vehicle.
  2. 2. The method of claim 1, wherein acquiring the obstacle information outside the vehicle and the intent of the occupant in the vehicle comprises: acquiring visual information of an obstacle at the rear side of a vehicle through a camera, acquiring distance information and speed information of the obstacle through a radar sensor, and identifying the type of the obstacle based on the visual information and the speed information, wherein the distance information is the relative distance between the obstacle and the target vehicle door; And acquiring the holding strength of the passenger in the vehicle and the pulling force along the opening direction of the vehicle door through a pressure sensor on a door handle of the target vehicle door, and determining the intention of the passenger in the vehicle based on the holding strength and the pulling force.
  3. 3. The method of claim 2, wherein determining the intent of the occupant in the vehicle based on the grip strength and the pull force comprises: judging whether the passenger in the vehicle has a door opening intention or not based on the holding strength; And if the passenger in the vehicle has the door opening intention, extracting the maximum change rate of the pulling force, and determining the door opening intention urgent degree of the passenger in the vehicle based on the maximum change rate of the pulling force, wherein the door opening intention urgent degree at least comprises a low urgent degree, a medium urgent degree and a high urgent degree.
  4. 4. The method of claim 2, wherein determining a risk level based on the obstacle information comprises: If the obstacle is a dynamic obstacle, calculating the shortest collision time of the dynamic obstacle and the target vehicle door based on the distance information and the speed information of the dynamic obstacle, and determining the risk weight of the dynamic obstacle based on the shortest collision time; if the obstacle is a static obstacle, determining a risk weight of the static obstacle based on an obstacle type of the static obstacle; a risk level is determined based on the risk weights of the dynamic obstacle and/or the static obstacle.
  5. 5. The method of claim 3, wherein generating the control command for the target door based on the risk level and the intent of the occupant in the vehicle comprises: If the risk level is low risk, generating a first control instruction when the door opening intention emergency degree is low emergency degree, and generating a second control instruction when the door opening intention emergency degree is medium emergency degree or high emergency degree, wherein the first control instruction is to apply original door mechanical opening resistance when a door is opened; If the risk level is the medium risk, a third control instruction is generated, wherein the third control instruction is to apply reverse resistance with a first preset magnitude when the vehicle door is opened, and trigger a second visual prompt, a first touch prompt and a first auditory prompt; And generating a fourth control instruction if the risk level is high risk, wherein the fourth control instruction is to apply a second preset resistance when the vehicle door is opened, trigger a third visual prompt, a second touch prompt and a second auditory prompt, the second preset magnitude is positively related to the risk weight of the high risk, and the second touch prompt is to start vibration according to a second frequency and a second amplitude through a vibration motor arranged in a handle of the target vehicle door.
  6. 6. The method of claim 1, further comprising, after obtaining the door lock state of the target door and the gear information and state information of the vehicle: If the target car door is in a locking state, generating a fifth control instruction to prohibit the target car door from being opened, wherein the fifth control instruction is used for triggering a fourth visual prompt and a third audible prompt; If the target vehicle door is in an unlocking state and the gear information of the vehicle is a non-target gear, generating a sixth control instruction to prohibit the target vehicle door from being opened, wherein the sixth control instruction is to apply a third preset resistance and trigger a fifth visual prompt and a fourth audible prompt; And if the target vehicle door is in an unlocking state, the gear information of the vehicle is in a target gear, and the vehicle is in a non-parking state, generating a seventh control instruction, wherein the seventh control instruction is to apply a fourth preset resistance and trigger a sixth visual prompt and a fifth audible prompt.
  7. 7. A method according to claim 3, further comprising: When the vehicle is detected to meet any emergency condition, a forced unlocking instruction and a vehicle door contact limiting instruction are generated, wherein the emergency condition comprises that the door opening intention emergency degree of an occupant in the vehicle is high emergency degree, the duration exceeds a time threshold, the distance between the obstacle and the target vehicle door is smaller than a distance threshold, an air bag deflagration signal, a smoke detector alarm signal or a vehicle body acceleration signal is detected, the number of times that the occupant in the vehicle pulls the same door handle in preset time is larger than or equal to preset times, and the pulling force is larger than or equal to a force threshold.
  8. 8. A multi-modal awareness based vehicle door collision avoidance decision system, comprising: The information acquisition module is used for acquiring the door lock state of the target door and gear information and state information of the vehicle; The multi-mode sensing module is used for acquiring the information of the obstacle outside the vehicle and the intention of the passenger in the vehicle if the target vehicle door is in an unlocking state, the gear information of the vehicle is in a target gear and the vehicle is in a parking state; And the decision module is used for determining a risk level based on the obstacle information and generating a control instruction of the target vehicle door based on the risk level and the intention of the passenger in the vehicle.
  9. 9. A vehicle comprising a processor and a memory, the memory storing computer executable instructions executable by the processor, the processor executing the computer executable instructions to implement the steps of the method of any one of claims 1 to 7.
  10. 10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, performs the steps of the method of any of the preceding claims 1 to 7.

Description

Multi-mode perception-based vehicle door anti-collision decision method and system Technical Field The invention relates to the technical field of automobile electronics and active safety, in particular to a vehicle door anti-collision decision method and system based on multi-mode perception. Background Currently, door collision avoidance techniques rely primarily on distance sensing based electronic sensing type restraint systems. The existing system detects the distance between an obstacle and a vehicle door through an ultrasonic radar or a millimeter wave radar, and adopts a fixed threshold value to judge whether to trigger the limitation of the vehicle door. And part of the system also links the vehicle door limit and the collision sensor, and automatically removes the limit when collision occurs, so as to ensure emergency escape. In addition, although existing systems attempt to detect the external environment of the vehicle door in real time by using sensing devices such as cameras and radars, decision basis of the systems is single, and mainly focuses on physical distance or existence of obstacles, so that the intelligent degree is low, and the system cannot adapt to complex and changeable traffic environments and passenger behaviors. Disclosure of Invention Therefore, the invention aims to provide a vehicle door anti-collision decision method and system based on multi-mode sensing, so that the intelligent degree of vehicle door anti-collision decision is improved, the safety is improved, and unnecessary risks caused by vehicle door collision are avoided. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: the invention provides a vehicle door anti-collision decision method based on multi-mode perception, which comprises the steps of obtaining a door lock state of a target vehicle door and gear information and state information of a vehicle, obtaining obstacle information outside the vehicle and intention of an occupant in the vehicle if the target vehicle door is in an unlocking state, the gear information of the vehicle is in a target gear and the vehicle is in a parking state, determining a risk level based on the obstacle information, and generating a control instruction of the target vehicle door based on the risk level and the intention of the occupant in the vehicle. The method comprises the steps of acquiring visual information of an obstacle behind a vehicle side through a camera, acquiring distance information and speed information of the obstacle through a radar sensor, and identifying the type of the obstacle based on the visual information and the speed information, wherein the distance information is the relative distance between the obstacle and a target vehicle door, acquiring the holding strength of the occupant in the vehicle and the pulling strength along the opening direction of the vehicle door through a pressure sensor on a door handle of the target vehicle door, and determining the intention of the occupant in the vehicle based on the holding strength and the pulling strength. Optionally, determining the intention of the occupant in the vehicle based on the holding power and the pulling force comprises judging whether the occupant in the vehicle has an intention to open the door based on the holding power, extracting the maximum change rate of the pulling force if the occupant in the vehicle has the intention to open the door, and determining the urgent degree of the intention to open the door of the occupant in the vehicle based on the maximum change rate of the pulling force, wherein the urgent degree of the intention to open the door at least comprises a low urgent degree, a medium urgent degree and a high urgent degree. Optionally, determining the risk level based on the obstacle information includes calculating a shortest collision time of the dynamic obstacle with the target vehicle door based on the distance information and the speed information of the dynamic obstacle if the obstacle is the dynamic obstacle, determining a risk weight of the dynamic obstacle based on the shortest collision time, determining a risk weight of the static obstacle based on an obstacle type of the static obstacle if the obstacle is the static obstacle, and determining the risk level based on the risk weights of the dynamic obstacle and/or the static obstacle. Optionally, generating a control command of a target vehicle door based on the risk level and the intention of an occupant in the vehicle, wherein the control command comprises generating a first control command when the door opening intention urgency is low and generating a second control command when the door opening intention urgency is medium or high if the door opening intention urgency is low if the risk level is low; If the risk level is the middle risk, a third control instruction is generated, wherein the third control instruction is to apply reverse resistance of a first preset m